ToTo Man
18-07-2018, 18:58
While I haven’t posted much on this forum, I have followed its threads with great interest for many years, especially those related to vintage British speakers. I am especially thankful to Ken (@Qwin) who was very helpful and patient with me as I sought his advice on re-capping a pair of Celestion Ditton 66 speakers a few years ago. Before I get to the crux of this post, please allow me to provide some background info.
I got into this audio hobby when I was 16 (I’m now 32) and thanks to sites like this, PFM and AK, I quickly became obsessed with vintage gear, especially loudspeakers, where the bigger and more “full-range” the better! As I progressed up the ladder through Goodmans, IMF, Celestion and Tannoy, I soon found that buying kit was much easier than selling it, at least from an emotional standpoint, especially when the stuff I was buying was so good. The house soon became a shrine to vintage hifi, much to the dismay of its other inhabitants!
After 16 years in the game, it has only been in the last year that I have become truly content and settled with my system, the focal point of which are Tannoy speakers that I doubt anyone else has in my particular configuration. After owning Lancaster Monitor Gold 15’s and Lockwood Major 15’s, I have finally settled on Monitor Gold 12’s in Edinburgh cabinets as these give me the flattest frequency response above 1kHz and do not require the use of DSP to make them behave linearly in my room. They provide the clearest and most transparent sound I have ever heard from a pair of loudspeakers, and it is very unlikely that anything else will displace them as my main reference speakers (unless I win a considerable sum on the lottery!). That said, I still like to keep a few other bits of kit around and ever the hoarder, I still have a couple of projects I need to finish. One of these is (another) pair of Celestion Ditton 66.
Warning: what follows is an attempt to describe a fairly complex predicament so it may take two or three read throughs to make sense!
Confession time; there are three pairs of 66’s in the house! My 1st pair I bought locally 13 years ago and is what started my love affair with this speaker. They are a little cosmetically challenged, but for the bargain price of £140 I couldn’t complain. My 2nd pair was significantly more expensive (hint: they cost me a shade more than what I believe Marco shelled out for his!). I tried to justify it by the fact that they had wooden fronts and were in utterly immaculate condition. My 3rd pair was unexpected. I spotted a pair of minty black-fronted cabinets on eBay for a reasonable price and decided to buy them, initially intending to transplant the drivers and XOs from my first pair into these. Greed got the better of me however and I decided instead to source all of the parts required to assemble a 3rd pair of working 66’s. This is the pair that Ken helped me re-cap.
My 1st and 2nd pairs still contain their stock crossovers. My 1st pair sounded fine when I bought them and still do, though I know they would almost certainly benefit from a re-cap. One MD500 driver in my 2nd pair (wood-fronted 66's) developed buzz/distortion very soon after I bought them, and on the advice of the seller I sent it to his tech to repair, however it came back sounding totally different but still had distortion! The project was thus put on hold until I sourced a replacement mid driver and found the time and energy to remove the crossovers and re-cap them. The time to finish said project has finally come…
One thing that’s always bugged me about vintage speakers is their tendency not to be acoustically matched. Sometimes the mismatch is subtle and can go unnoticed but sometimes it’s really obvious, to my ears anyway. Now that I have the ability to acoustically measure driver frequency response and distortion, I spent last weekend measuring the raw response of all the spare 66 drivers I have stored away, and I also pulled all of the drivers from my 1st and 2nd pair of 66's and measured them too. It was an exhausting weekend!
The T1600 and T2619 bass drivers all match perfectly in terms of FR and are within 1dB in terms of efficiency, - so far so good!:
https://farm2.staticflickr.com/1784/43444685622_9b6fc1bb12_b.jpg
The MF500 and MD500 units match each other perfectly in terms of FR and are within 0.5dB in terms of efficiency, - even better!. The MD version has stronger output between 1.75kHz and 5kHz, useful to know:
https://farm2.staticflickr.com/1761/43492389831_d13b3fc47d_b.jpg
The HF2000 units aren’t as good. From a sample of 13 units I can only find two that are within 2dBs of each other for most of the range. Some are out by as much as 5 or 6dBs!!! On top of that, many of their FR curves are particular uneven with big dips and peaks:
https://farm2.staticflickr.com/1803/43444687312_1b1c515378_b.jpg
https://farm2.staticflickr.com/1826/43444688282_571cd5baaa_b.jpg
https://farm2.staticflickr.com/1805/43444688102_46d23dd1b1_b.jpg
https://farm2.staticflickr.com/1805/43444687972_024a8c2211_b.jpg
https://farm1.staticflickr.com/847/43444687852_102b2a90e6_b.jpg
https://farm2.staticflickr.com/1806/43444687602_db27391ae0_b.jpg
https://farm1.staticflickr.com/916/43444687472_871db2743c_b.jpg
https://farm2.staticflickr.com/1829/43444686912_d8624a7284_b.jpg
The best matched pair is obtained by partnering unit #6 with unit #8:
https://farm2.staticflickr.com/1830/43444687012_722ebe6e79_b.jpg
When I assembled my last pair of 66’s I did not have access to measuring equipment, so I matched the speakers by ear. As this FR graph shows, I got a pretty good match (luck or golden ears?!):
https://farm2.staticflickr.com/1801/43444688882_319e01f6e8_b.jpg
However the graph shows that the high frequencies are too loud compared to the mids when measured at 1m on-axis with the tweeter. I suspect this is partly due to 1 metre being too close a distance to measure from, due to either diffraction from the top lip of the cabinet and/or cancellations between the HF2000 and MF500. I’m not an expert, it’s just a hunch. The response does become flatter when I measure from my listening seat:
https://farm2.staticflickr.com/1765/43492390901_b2d96bacbf_b.jpg
Raising the speakers up off the floor also smooths out the mid/treble balance a little. This is the nearfield response with the tweeter at different heights above the mic. The highest I’ve tried so far is 200mm:
https://farm2.staticflickr.com/1826/43444688612_4336bb918a_b.jpg
The graph above is a bit crowded so here is the nearfield response at 200mm height only:
https://farm2.staticflickr.com/1830/43444686082_0946fddafa_b.jpg
Learning from the experience of assembling my last pair of 66’s, my plan for my wood-fronted 66's is to select the drive units that are the flattest and closest match in terms of raw FR and SPL. By doing this, I would be ignoring the drivers DCRs. However I have been told that the DCR is important to consider because it affects the crossover slopes and can either contribute to a peak, dip, or flat response at the crossover point. The DCRs of the HF2000s I have range from 3.4 to 4.7 ohms.
So, my first question is, when selecting the drivers to go into my 66's, how much importance do I give to the drivers raw FR and SPL efficiency and how much importance do I give to its DCR?
Second, I was under the impression that the black-fronted 66 came with point-to-point crossover board, T1600, MF500 and 30uf for the MF circuit, and that the wood-fronted 66 came with PCB crossover board, T2619, MD500 and 24uf for the MF circuit. However, upon removing all of the drivers to measure them, I discovered that my wood-fronted 66 has point-to-point crossover board, T2619 and MD500, but the left crossover has 30uf for the MF circuit (it has a big red cap which I presume is 24uf plus a little 6uf elcap) while the right crossover just has a 24uf elcap for the MF circuit. So it seems like Celestion were experimenting with different values of caps on the MD500 driver?
I want to use the MD500 in my wood-fronted 66’s because it has stronger output between 1.75kHz and 5kHz, and my ears tell me my black-fronted 66’s could benefit from more presence in this region. I was planning to use 24uf in the MF circuit (24.6uf to be exact, made up from 23.6uf Mundorf ECap and 1uf Ansar), but now that I see that Celestion sometimes used 24uf and other times used 30uf in the MF circuit with the MD500, I’m no longer so sure what to use. Can someone please clarify this?
Finally, when I assembled my last pair of 66’s, I didn’t pay much attention to the quality of the hookup wire and just used some basic stuff I bought from Maplins a few years ago. I used this for the jumpers between the solder points on the board and also for the wires to the drivers because it was easy to work with (I am dismal at soldering!). I don’t know its spec, it’s quite thin but it’s thicker than the stock wire:
https://farm2.staticflickr.com/1789/43492389731_766c4c9e59_b.jpg
If I’m going to the effort of re-capping my wood-fronted 66’s I suppose I ought to use decent hookup wire. What would you recommend? Or do you think what I used in my other 66’s is fine?
Thanks for taking the time to read my rambling post, I hope you were able to make it to the end without losing the will to live… I almost did! It’s taken me a full two days to write this as I have cerebral palsy and have only the use of one finger with which to type!
I got into this audio hobby when I was 16 (I’m now 32) and thanks to sites like this, PFM and AK, I quickly became obsessed with vintage gear, especially loudspeakers, where the bigger and more “full-range” the better! As I progressed up the ladder through Goodmans, IMF, Celestion and Tannoy, I soon found that buying kit was much easier than selling it, at least from an emotional standpoint, especially when the stuff I was buying was so good. The house soon became a shrine to vintage hifi, much to the dismay of its other inhabitants!
After 16 years in the game, it has only been in the last year that I have become truly content and settled with my system, the focal point of which are Tannoy speakers that I doubt anyone else has in my particular configuration. After owning Lancaster Monitor Gold 15’s and Lockwood Major 15’s, I have finally settled on Monitor Gold 12’s in Edinburgh cabinets as these give me the flattest frequency response above 1kHz and do not require the use of DSP to make them behave linearly in my room. They provide the clearest and most transparent sound I have ever heard from a pair of loudspeakers, and it is very unlikely that anything else will displace them as my main reference speakers (unless I win a considerable sum on the lottery!). That said, I still like to keep a few other bits of kit around and ever the hoarder, I still have a couple of projects I need to finish. One of these is (another) pair of Celestion Ditton 66.
Warning: what follows is an attempt to describe a fairly complex predicament so it may take two or three read throughs to make sense!
Confession time; there are three pairs of 66’s in the house! My 1st pair I bought locally 13 years ago and is what started my love affair with this speaker. They are a little cosmetically challenged, but for the bargain price of £140 I couldn’t complain. My 2nd pair was significantly more expensive (hint: they cost me a shade more than what I believe Marco shelled out for his!). I tried to justify it by the fact that they had wooden fronts and were in utterly immaculate condition. My 3rd pair was unexpected. I spotted a pair of minty black-fronted cabinets on eBay for a reasonable price and decided to buy them, initially intending to transplant the drivers and XOs from my first pair into these. Greed got the better of me however and I decided instead to source all of the parts required to assemble a 3rd pair of working 66’s. This is the pair that Ken helped me re-cap.
My 1st and 2nd pairs still contain their stock crossovers. My 1st pair sounded fine when I bought them and still do, though I know they would almost certainly benefit from a re-cap. One MD500 driver in my 2nd pair (wood-fronted 66's) developed buzz/distortion very soon after I bought them, and on the advice of the seller I sent it to his tech to repair, however it came back sounding totally different but still had distortion! The project was thus put on hold until I sourced a replacement mid driver and found the time and energy to remove the crossovers and re-cap them. The time to finish said project has finally come…
One thing that’s always bugged me about vintage speakers is their tendency not to be acoustically matched. Sometimes the mismatch is subtle and can go unnoticed but sometimes it’s really obvious, to my ears anyway. Now that I have the ability to acoustically measure driver frequency response and distortion, I spent last weekend measuring the raw response of all the spare 66 drivers I have stored away, and I also pulled all of the drivers from my 1st and 2nd pair of 66's and measured them too. It was an exhausting weekend!
The T1600 and T2619 bass drivers all match perfectly in terms of FR and are within 1dB in terms of efficiency, - so far so good!:
https://farm2.staticflickr.com/1784/43444685622_9b6fc1bb12_b.jpg
The MF500 and MD500 units match each other perfectly in terms of FR and are within 0.5dB in terms of efficiency, - even better!. The MD version has stronger output between 1.75kHz and 5kHz, useful to know:
https://farm2.staticflickr.com/1761/43492389831_d13b3fc47d_b.jpg
The HF2000 units aren’t as good. From a sample of 13 units I can only find two that are within 2dBs of each other for most of the range. Some are out by as much as 5 or 6dBs!!! On top of that, many of their FR curves are particular uneven with big dips and peaks:
https://farm2.staticflickr.com/1803/43444687312_1b1c515378_b.jpg
https://farm2.staticflickr.com/1826/43444688282_571cd5baaa_b.jpg
https://farm2.staticflickr.com/1805/43444688102_46d23dd1b1_b.jpg
https://farm2.staticflickr.com/1805/43444687972_024a8c2211_b.jpg
https://farm1.staticflickr.com/847/43444687852_102b2a90e6_b.jpg
https://farm2.staticflickr.com/1806/43444687602_db27391ae0_b.jpg
https://farm1.staticflickr.com/916/43444687472_871db2743c_b.jpg
https://farm2.staticflickr.com/1829/43444686912_d8624a7284_b.jpg
The best matched pair is obtained by partnering unit #6 with unit #8:
https://farm2.staticflickr.com/1830/43444687012_722ebe6e79_b.jpg
When I assembled my last pair of 66’s I did not have access to measuring equipment, so I matched the speakers by ear. As this FR graph shows, I got a pretty good match (luck or golden ears?!):
https://farm2.staticflickr.com/1801/43444688882_319e01f6e8_b.jpg
However the graph shows that the high frequencies are too loud compared to the mids when measured at 1m on-axis with the tweeter. I suspect this is partly due to 1 metre being too close a distance to measure from, due to either diffraction from the top lip of the cabinet and/or cancellations between the HF2000 and MF500. I’m not an expert, it’s just a hunch. The response does become flatter when I measure from my listening seat:
https://farm2.staticflickr.com/1765/43492390901_b2d96bacbf_b.jpg
Raising the speakers up off the floor also smooths out the mid/treble balance a little. This is the nearfield response with the tweeter at different heights above the mic. The highest I’ve tried so far is 200mm:
https://farm2.staticflickr.com/1826/43444688612_4336bb918a_b.jpg
The graph above is a bit crowded so here is the nearfield response at 200mm height only:
https://farm2.staticflickr.com/1830/43444686082_0946fddafa_b.jpg
Learning from the experience of assembling my last pair of 66’s, my plan for my wood-fronted 66's is to select the drive units that are the flattest and closest match in terms of raw FR and SPL. By doing this, I would be ignoring the drivers DCRs. However I have been told that the DCR is important to consider because it affects the crossover slopes and can either contribute to a peak, dip, or flat response at the crossover point. The DCRs of the HF2000s I have range from 3.4 to 4.7 ohms.
So, my first question is, when selecting the drivers to go into my 66's, how much importance do I give to the drivers raw FR and SPL efficiency and how much importance do I give to its DCR?
Second, I was under the impression that the black-fronted 66 came with point-to-point crossover board, T1600, MF500 and 30uf for the MF circuit, and that the wood-fronted 66 came with PCB crossover board, T2619, MD500 and 24uf for the MF circuit. However, upon removing all of the drivers to measure them, I discovered that my wood-fronted 66 has point-to-point crossover board, T2619 and MD500, but the left crossover has 30uf for the MF circuit (it has a big red cap which I presume is 24uf plus a little 6uf elcap) while the right crossover just has a 24uf elcap for the MF circuit. So it seems like Celestion were experimenting with different values of caps on the MD500 driver?
I want to use the MD500 in my wood-fronted 66’s because it has stronger output between 1.75kHz and 5kHz, and my ears tell me my black-fronted 66’s could benefit from more presence in this region. I was planning to use 24uf in the MF circuit (24.6uf to be exact, made up from 23.6uf Mundorf ECap and 1uf Ansar), but now that I see that Celestion sometimes used 24uf and other times used 30uf in the MF circuit with the MD500, I’m no longer so sure what to use. Can someone please clarify this?
Finally, when I assembled my last pair of 66’s, I didn’t pay much attention to the quality of the hookup wire and just used some basic stuff I bought from Maplins a few years ago. I used this for the jumpers between the solder points on the board and also for the wires to the drivers because it was easy to work with (I am dismal at soldering!). I don’t know its spec, it’s quite thin but it’s thicker than the stock wire:
https://farm2.staticflickr.com/1789/43492389731_766c4c9e59_b.jpg
If I’m going to the effort of re-capping my wood-fronted 66’s I suppose I ought to use decent hookup wire. What would you recommend? Or do you think what I used in my other 66’s is fine?
Thanks for taking the time to read my rambling post, I hope you were able to make it to the end without losing the will to live… I almost did! It’s taken me a full two days to write this as I have cerebral palsy and have only the use of one finger with which to type!